[llvm] [ExpandVectorPredication] Expand vp.load.ff. (PR #154440)

[Luke Lau via llvm-commits]

================
@@ -465,6 +465,21 @@ CachingVPExpander::expandPredicationInMemoryIntrinsic(IRBuilder<> &Builder,
         VPI.getName());
     break;
   }
+  case Intrinsic::vp_load_ff: {
+    // Mask out all but the first lane.
+    Value *NewMask = ConstantInt::getFalse(MaskParam->getType());
+    NewMask = Builder.CreateInsertElement(
+        NewMask, ConstantInt::getTrue(MaskParam->getType()->getScalarType()),
+        (uint64_t)0);
----------------
lukel97 wrote:

Why do we need the (uint64_t) cast?

https://github.com/llvm/llvm-project/pull/154440